- Email: [email protected]
Evaluation of diet quality of the elderly and associated factors
Accepted Manuscript Title: EVALUATION OF DIET QUALITY OF THE ELDERLY AND ASSOCIATED FACTORS Authors: Dalila Pinto de Souza Fernandes, Maria Sˆonia Lopes Duarte, Milene Cristine Pessoa, Sylvia do Carmo Castro Franceschini, Andr´eia Queiroz Ribeiro PII: DOI: Reference:
S0167-4943(17)30230-3 http://dx.doi.org/doi:10.1016/j.archger.2017.05.006 AGG 3482
To appear in:
Archives of Gerontology and Geriatrics
Received date: Revised date: Accepted date:
17-10-2016 6-5-2017 9-5-2017
Please cite this article as: Pinto de Souza Fernandes, Dalila, Duarte, Maria Sˆonia Lopes, Pessoa, Milene Cristine, Franceschini, Sylvia do Carmo Castro, Ribeiro, Andr´eia Queiroz, EVALUATION OF DIET QUALITY OF THE ELDERLY AND ASSOCIATED FACTORS.Archives of Gerontology and Geriatrics http://dx.doi.org/10.1016/j.archger.2017.05.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
EVALUATION OF DIET QUALITY OF THE ELDERLY AND ASSOCIATED FACTORS Dalila Pinto de Souza Fernandes1 Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil (32) 98428-5058 [email protected] Maria Sônia Lopes Duarte1 Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil (31) 3899-3741 [email protected] Milene Cristine Pessoa2 Department of Nutrition, Nursing school, Federal University of Minas Gerais, Av. Prof. Alfredo Balena, 190, Santa Efigênia, Belo Horizonte, MG, 30130-100, Brazil (31) 3409-9179 [email protected] Sylvia do Carmo Castro Franceschini1 Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil (31) 3899-3743 [email protected] Andréia Queiroz Ribeiro1 Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil (31) 3899-1271 [email protected]
1
2
Federal University of Viçosa Federal University of Minas Gerais
Please address all correspondence to: Dalila Pinto de Souza Fernandes Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil e-mail: [email protected] Tel: +55 – 32 – 98428-5058
ABSTRACT Background: Observational studies suggest healthy dietary patterns are associated with risk reduction and better control of various chronic diseases. However, few Brazilian studies have focused on evaluating the quality of the elderly diet and its relationship with diseases. This study aimed to estimate the association between diet quality and socioeconomic factors, health and nutrition of the elderly. Methods: This is a cross-sectional population-based study whose target population were non-institutionalized elderly residents in the city of Viçosa, Brazil. Anthropometric, socioeconomic, health conditions, lifestyle and food consumption variables were obtained from a semi-structured questionnaire. The quality of the diet was assessed by the revised Healthy Eating Index classified into tertiles, considering the first tertile as "Poor diet quality," the second as "Intermediate diet quality" and the third as "Better diet quality." To identify factors independently associated with diet quality model, the works used multinomial logistic regression. Results: In the results of the multivariate analysis, the factors independently associated with "better diet quality" included female gender, higher education, history of one to five medical visits in the past year, history of diabetes mellitus, dyslipidemia and the use of polypharmacy. Discussion: Our results show that most seniors need to improve the quality of their diet and those of male gender with no or little education, and those who do not seek medical services constitute the group that needs attention concerning the measures to improve the quality of their diet.
Keywords: Diet; Diet quality; Food consumption; Elderly; Elderly health.
Highlights The diet quality of brazilian elderly requires improvement. Elderly women and those with some education level presented better diet quality. Elderly with diabetes mellitus and dyslipidemia presented better diet quality.
Elderly who do not seek health services are the groups that needs more attention.
EVALUATION OF DIET QUALITY OF THE ELDERLY AND ASSOCIATED FACTORS 1. Introduction
Eating habits play a unique role in health, as the combination of food consumed over time throughout the life cycle can be determinant of the occurrence of various diseases (Hiza et al., 2013). Results from observational studies suggest that healthy dietary patterns are associated with risk reduction and better control of several chronic diseases such as obesity, dyslipidemia, hypertension and diabetes mellitus (Lopes et al., 2005; Kant, 2010; Giuli et al., 2012; Hiza et al., 2013). Among the elderly, nutrition becomes particularly essential for the good maintenance of acceptable health standards and functional capacity (Silva, Pedraza & Menezes, 2015). In order to properly estimate the relationship between diet and morbidities, specific tools have been developed to assess food consumption (Fisberg et al., 2004). Because of the relationship between food, the variety of factors associated with its intake and the prevalence of certain diseases that increase with advancing age, we consider of great importance to conduct the diet evaluation globally since it would better reflect the complexity of this relationship (Louzada et al., 2012). In this context, the Brazilian Healthy Eating Index–Revised (BHEI-R) (Previdelli et al., 2011)
was
proposed as dietary assessment tool, wich is based on the food guide recommendations for the Brazilian population (BRASIL, 2008) and the Healthy Eating Index of 2005 (Guenther, Reedy & Krebs-Smith, 2008a). Few Brazilian studies published in scientific literature have focused on evaluating the diet quality of the elderly. Among the published studies, the authors have observed an association between a poorer diet quality and low education, low income, lack of appetite, difficulty in purchasing and preparing food (Viebig et al., 2009) in addition to smoking, consumption of sugary drinks and alcohol (Assumpção et al., 2014). On the other hand, the elderly with nutritional counseling, non-smokers and those who practice physical activity have shown better diet quality (Gadenz, Benvegnú, 2013; Assumpção et al., 2014). However, there are few studies with Brazilian elderly that assessed dietary intake through the BHEI-R and that took chronic diseases that may be related to food into consideration. In this sense, the present study aimed to estimate
the association between diet quality and specific socioeconomic factors, health and nutrition of the elderly.
2. Methodology 2.1. Designing the study This is a cross-sectional population-based study, which integrates the research project "Health conditions, nutrition and medication use by the elderly in Viçosa (MG): a population-based survey" conducted from June to December 2009. This project was approved by the Ethics Committee on Human Research of the Federal University of Viçosa (Official letter No. 27/2008/CEP/UFV). The evaluation of the elderly was carried out only after all parties signed the Free and Informed Consent Term (FICT).
2.2. Target population and sample The study’s target population consisted of seniors aged over 60 years, who are not institutionalized, residents in rural and urban areas of Viçosa, which corresponded to 7980 elderly in 2008 (study period). Those data were the basis for the calculation of the sample. The sample size was calculated considering a confidence level of 95%, estimated prevalence of 50% for different outcomes of interest to the larger project, 4% tolerated error and 20% increase to cover losses. From an initial study sample of 670 participants, 50 were excluded because there were losses due to subject refusal to participate (3.6%), unavoidable reasons that interviews could not be conducted such as in situations where randomly selected individuals had died (1.3%), had an address that could not be located in the city (1.2%) or had moved to other locations which were either difficult to find or in other municipalities (1,2%), in addition to missing data from the recall of habitual consumption (0,15%). Those participants excluded were no different from the included ones regarding gender and age range (Nascimento et al., 2012). Thus, excluding losses, the final sample was of 620 elderly, what represents a response rate of 93.0%.
2.3. Data collection
The interviews were previously scheduled, they were performed at each elderly’s home and conducted by pairs of trained interviewers. Up to three attempts to visit the interviewee’s home were performed before the person was considered a loss. The data collection instrument consisted of a semi-structured questionnaire with sociodemographic variables, health conditions, lifestyle and food consumption. To obtain the food consumption data we have applied a standard recall of habitual consumption using the 'multi-pass' technique (Guenther, P.M., et al., 1995). The contents of the recall were reviewed prior to data entry. Therefore, we have performed a standardization of preparations and percentage of salt and oil (Pinheiro et al., 2005; Araújo, Guerra, 2007; TACO, 2011). Added sugars included all sugars added to foods during processing, preparation or consumption. The amount of added sugar in each food was standardized based on the Table for Evaluation of Food Consumption in Household Measurements (Pinheiro et al., 2005), in recipes and on foods labels. Later, the data were entered in Dietpro® version 5i software, and it were performed quality control of the typed data. The study has also measured the interviewee’s waist circumference (WHO, 1995).
2.4. Study variables The Brazilian Healthy Eating Index–Revised (BHEI-R), which evaluates the diet quality and was obtained from the information from the recall of habitual consumption, was used as dependent variable. This index was revised by Previdelli et al (2011) and validated for Brazilian population by Andrade et al (2013). In relation to the total score, the maximum value of the total BHEI-R is 100 points. Values closer to the maximum score represent better diet quality. Because the index is designed to reflect different aspects of the diet, there is no classification in adequate and inadequate considering the total score (Guenther, Reedy & Krebs-Smith, 2008a; Guenther, Reedy & Krebs-Smith, 2008b). Therefore, the total BHEI-R was ranked into tertiles, considering the first tertile as "Poor diet quality," the second as "Intermediate diet quality" and the third as "Best diet quality." The independent variables in this study included: Socioeconomic: gender (male, female); age range (60-74 years, 75 years and older); education (never studied; early grades of elementary school; final grades of elementary school or higher); individual income (quartiles) and cohabitation (living alone, living together).
Health conditions: number of medical appointments (none; 1-5; 6 or more) during the year before the interview and number of hospitalizations (none, 1 or more) during the year before the interview. Functional capacity was assessed by questions about ability to perform 14 basic and instrumental activities of daily living, categorized into: no struggle, little struggle, great struggle, unable to do it and does not do it (Nascimento et al., 2012). In the analysis, we have considered the variables difficulty to feed and difficult to prepare food, as well as functional capacity defined as inadequate, when the elderly reported difficulty performing seven or more activities, or when self-assessed as unable to perform three or more activities; and otherwise as appropriate. We have also evaluated the polypharmacy (yes, no), defined as the use of five or more drugs over the 15 days preceding the interview2 (Beloosesky et al., 2013). As for morbidities, information about the history of chronic diseases were obtained from the following question: "Have you ever in life had a doctor or other health professional said that you had or have had any of the following diseases?". Diseases of interest to the study included diabetes mellitus; arthrosis, arthritis or rheumatism; heart attack; asthma or bronchitis; depression; hearing problems; high pressure; angina; stroke; vision problems; osteoporosis; dyslipidemia; kidney disease and cancer. For the analysis we used the total number of disorders and history of diabetes mellitus, dyslipidemia and high blood pressure.
Lifestyle: alcohol consumption (yes; no but has drunk before; never drank) and smoking (yes; no but has smoked before; never smoked).
Anthropometric measure: we have measured the waist circumference (WC) using flexible and inelastic tape (TBW®, São Paulo, Brazil) 1.80m long and 0.1mm accuracy. This variable was measured with the elderly in the standing position, the height of the midpoint between the last rib and the iliac crest, during expiration, according to the protocol of the World Health Organization, 1995 (WHO, 1995).
2.5. Data analysis For this study, we proceeded the descriptive analysis of variables of interest, based on the frequency distribution and estimates of central tendency and dispersion. The normal distribution of the BHEI-R values between the categories of the variables was assessed using the Shapiro-Wilk test.
Multinomial logistic regression analysis was used to obtain estimates for Odds Ratios and 95% confidence intervals of the association between the independent variables of interest in the study and the diet quality. This method of regression is used when the dependent variable has more than two categories, each being compared to a reference category. In this study the reference category was the lowest tertile values of BHEI-R, which was called "Poor diet quality." The variables that were associated with the dependent variable in univariate step with p<0.20 were included in the multivariate multinomial logistic regression model. The variables that were associated with the dependent variable at the level of p<0.05 remained in the final model. All analyzes were performed using STATA software (Stata Corp. College Station, United States) version 13.0, adopting α = 0.05 as statistical significance level for all comparisons.
3. Results
The sample consisted of 620 elderlies. According to the socioeconomic characteristics, most were female (53.2%), aged between 60 and 74 years (69.8%), studied until the early grades of elementary school (64.0%) and did not live alone (89.4%). About 38.0% of the elderly had individual monthly income between $ 847.88 and $ 952.71 (second quartile), whereas the lower limit corresponds to the minimum wage during the period of the study (Table 1). Regarding to health and food consumption, the sample characterization is shown in Table 2. We have identified that 48.3% of the elderly rated their health as fair, most (72.5%) had a history of one to five doctor appointments and 15.0% had a hospital history during the previous year. Functional capacity was classified as inappropriate in 16.0% of the elderly, and 8.6% reported difficulty to feed and 16.1% reported difficulty to prepare food. Polypharmacy was observed in 36.1% of seniors and 38,0% reported having five or more diseases. About 22.0% had a history of diabetes mellitus, 56.0% of dyslipidemia and 76.5% of high pressure. In relation to alcohol consumption, 33.8% admitted to drinking and just over half of the sample reported never having smoked. Approximately 13,0% of the sample reported some reduction in food intake over the three months prior to the survey. Waist circumference and total BHEI-R are shown in tertiles (Table 2). The results of the univariate analysis showed that the factors positively
associated with the tertile "Intermediate diet quality" included female gender, having history of diabetes mellitus and having waist circumference between 89,70 and 100,79 cm. For the tertile "Best diet quality," the positive and significant associations were female gender, having studied at least until early grades of elementary school, having made at least one medical visit in the past year, having history of five or more diseases, polypharmacy, history of diabetes mellitus and dyslipidemia, and having waist circumference between 59,67 and 70,05cm (Tables 3 and 4). The results of multivariate analysis allowed us to observe the factors which remained independently associated to the tertile "Intermediate diet quality" included female gender and having waist circumference between 89,70 and 100,70cm. On the other hand, the factors independently associated with tertile of "Best diet quality" were female gender, having higher education, having had five medical appointments in the past year, having history of diabetes mellitus, dyslipidemia and having practiced polypharmacy. The variable for number of disorders was not included in this multivariate analysis because the variables history of diabetes mellitus and history of dyslipidemia were included in the first (Table 5).
4. Discussion
The results indicate that the diet quality of the elderly requires improvement, considering that two-thirds had scores ranging between about 22 and 70 points. This finding is also reported by other authors (Ervin, Dye, 2008; Savoca et al., 2009; Cannella, Savina & Donini, 2009; Louzada et al., 2012; Assumpção et al., 2014). In this study, the goal was to investigate the association of socioeconomic variables, health status, use of health services, lifestyle and waist circumference with the elderlies’ tertiles for BHEI-R. Being female remained positively associated with tertiles "Intermediate diet quality" and "Best diet quality". Among the elderly, 55.6% belonged to the tertile of "Intermediate diet quality" and 60.9% to tertile "Best diet quality". International and national studies show that women generally have better diet quality than men(Louzada et al., 2012; Malta, Moura & Morais Neto, 2011; Hiza et al., 2013), which has been justified by the fact that they have a greater concern for their health, and with healthier eating habits (Francisco et al., 2015). Having education was positively associated with "Best diet quality", i.e. the elderly who have a healthier diet had a greater chance of having education compared
those in tertile "Poor diet quality". The strength of association was higher the higher the level of education. This association was also found in other studies (Fisberg et al., 2004; Ervin, Dye, 2008; Basiotis et al., 2002). Elderly American participants from the National Health and Nutrition Examination Survey 2003-2004 in which higher levels of education also showed better total score of HEI (Hiza et al., 2013). In a study with elderly in the Brazilian South, it was observed that those with higher education consume less carbohydrates and more foods rich in minerals and vitamins compared to those with less than 8 years of schooling, noting that the level of education is crucial to healthy food habits (Venturini et al., 2015). When the elderly showed one to five medical appointments during the year before the interview, this was positively associated with a better diet quality, as in a study with elderly in Southern Brazil (Gadenz, Benvegnú, 2013). There was also better diet quality among older people who reported a history of diabetes mellitus, dyslipidemia and use of polypharmacy. All these findings can be attributed to what is called reverse causality, as in this case, the diagnosis of a disease may raise a need for change in lifestyle in the individual, and may thus result in the adoption of healthier eating habits. The positive association between history of diabetes mellitus and "Best diet quality" probably results from the change in behavior, since diet is one of the key factors for glycemic control (Jenkins et al., 2008; Exebio et al., 2011). In a populationbased study for elderly people in São Paulo, we found similar results with better diet quality among those diagnosed with diabetes mellitus (Assumpção et al., 2014).
In
addition to diabetes mellitus, dyslipidemia and obesity are also common in this age group33, and the appropriate control and prevention of complications from these diseases are associated with healthy eating habits (Kant, 2010; Giuli et al., 2012). Our results, although consistent with the literature, have the limitation of not assessing whether the elderly with better diet quality had adequate control of the disease because no information was available for this evaluation. This aspect can be explored in future studies, especially those of longitudinal design to allow more evidence about the influence of the diet quality over the control of chronic non-communicable diseases in the elderly. Polypharmacy appears as a frequent practice among the elderly and its prevalence in the sample (36.1%) was higher than that observed in other Brazilian studies (5-27%) (Flores, Mengue, 2005; Loyola Filho et al., 2005; Flores, Benvegnú,
2008; Duarte et al., 2012; Gontijo et al., 2012) On the one hand, polypharmacy can be seen as an indicator of increased morbidity among the elderly, compared to its strong association with other diseases. In this study, its association with the "Best diet quality" may indicate that seniors using five or more medications are more concerned with healthy eating. Notwithstanding, there is an important concern about the influence of using multiple drugs in the elderly nutritional status. Certain studies have evaluated its association with nutritional factors and it is known that the use of multiple drugs, inappropriate prescribing and use of two or more drugs with the same pharmacological activity, among other factors, promote increased side effects and interactions that interfere with digestion, absorption and metabolism of nutrients (Faria, Franceschini & Ribeiro, 2010).
Additional studies are needed to verify the actual interference of
polypharmacy in the bioavailability of nutrients as well as longitudinal studies can help to elucidate the influence of polypharmacy in the nutritional status of the elderly. As the high concentration of serum lipids, abdominal obesity is a major risk factor for cardiovascular diseases (Rocha et al., 2013). As waist circumference (WC) is an indicator of central adiposity, it has been associated with the risk of morbidity and mortality in the elderly (Romaguera et al., 2011). In the present study, only the second tertile for WC values remained associated with "Intermediate quality diet", which also suggests a change in behavior. The fact that the WC tertile with the highest values is not associated with the "Best diet quality" can mean low sample power, given that the study was not specifically designed to evaluate this relationship. Some limitations should be considered when interpreting the findings of the present study. First, its cross-sectional design limits the interpretation of some associations found as a result of cause and effect relationship. However, the results are consistent with the literature. Second, some variables analized in this study were selfreported, however, studies in adults and the elderly attest the reliability of this information (Martin, et al., 2000; Lima-Costa, Peixoto & Firmo, 2004; Selem, et al., 2013). Third, the variable physical activity was not considered as a control, due to the lack of information about kind of activity, intensity and frequency. At this point of view, the relationship between physical activity practice and diet quality improvement should be adressed in future studies. Fourth, there are the limitations inherent to dietary surveys such as errors in the estimate of food portions and memory bias (Scagliusi, Lancha Júnior, 2003). One reservation should be made to the fact that we have used the recall of habitual consumption as consumer assessment tool, which despite not being
validated, allowed us to observe that the elderly tend to have monotonous diets as the difficulties to prepare food and even to eat increase (Venturini et al., 2015) and even due to income issues, which minimizes the potential bias in this case and justifies the use of this recall. Moreover, this recall provides a better estimate of food intake when compared to the Food Frequency Questionnaire, which has a limited number of food items. This study’s results show most seniors need to improve their diet quality and there is evidence of the association between female gender, having higher education, having history of diabetes mellitus, dyslipidemia, using polypharmacy and having the "Best quality diet". Considering some methodological differences between the studies, these results are similar to those observed in the elderly in developed countries and larger cities in Brazil. Elderly men, with any or little education and those who do not seek health services are the groups that needs more attention regarding to the measures to improve their diet quality.
Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest. Authors’ contributions Fernandes DPS worked in study design, analysis and interpretation of data, article writing and critical review of relevant intellectual content and final approval of the version to be published. Ribeiro AQ guided the design, collection and analysis and interpretation of data, and critical review of relevant intellectual content and final approval of the version to be published. Duarte MSL and Pessoa MC collaborated in the analysis and interpretation of data, article writing and critical review of relevant intellectual content; Franceschini SCC collaborated in the relevant critical review of the intellectual content.
Ethical The study was conducted according to the directives established in the Declaration of Helsinki, and all procedures involving human being were approved by the Research Ethics Committee of the Federal University of Viçosa, Brazil. All persons gave their informed consent prior to their inclusion in the study.
Acknowledgments The Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), (process n° 579255/2008-5) funded this project and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) provided Master’s degree scholarship. The authors thank the team of researchers who collected the data and the Grupo de Estudo sobre Envelhecimento, Nutrição e Saúde (GREENS) of the Department of Nutrition and Health of the Federal University of Viçosa.
References Andrade, S. C., Previdelli, Á. N., Marchioni, D. M. L., & Fisberg, R. M. (2013). Evaluation of the reliability and validity of the Brazilian Healthy Eating Index Revised. Revista de Saúde Pública, 47, 675-83. Araújo, M. O. D., & Guerra, T. M. M. (2007). Alimentos “Per capita”. (3th ed.). UFRN: Natal. Assumpção, D., Domene, S. M. Á., Fisberg, R. M., & Barros, M. B. A. (2014). Diet quality and associated factors among the elderly: a population-based study in Campinas, São Paulo State, Brazil. Cadernos de Saúde Pública, 30, 1680-1694. Basiotis, P. S. et al. The Healthy Eating Index: 1999-2000. (2002). U.S. Department of Agriculture, Center for Nutrition Policy and Promotion. CNPP-12. p. 1999–2000. Beloosesky, Y. et al. (2013). Rates, variability, and associated factors of polypharmacy in nursing home patients. Clinical Interventions in Aging, 8, 1585–1590. Brasil (2008) Guia alimentar para a população brasileira: promovendo à alimentação saudável. Secretaria de Atenção à Saúde, Coordenação-Geral da Política de Alimentação e Nutrição. Ministério da Saúde, Brasília. Cannella, C., Savina, C., & Donini, L. M. (2009). Nutrition, longevity and behavior. Archives of gerontology and geriatrics, 49, 19–27. Duarte, L. R. et al. (2012). Habits of medication consumption among elderly SUS and health insurance users. Cadernos de Saúde Coletiva, 20, 64–71. Ervin, R. B., & Dye, B. A. (2008). Eating Index scores among adults, 60 years of age and over, by sociodemographic and health characteristics: United States, 1999–2002. Advanced Data, 20, 1–16. Exebio, J. C. et al. (2011). Healthy Eating Index scores associated with symptoms of depression in Cuban-Americans with and without type 2 diabetes: a cross sectional study. Nutrition Journal, 10, 135-141.
Faria, M. Q., Franceschini, S. C. C., & Ribeiro, A. Q. (2010). Estado Nutricional e Uso de Medicamentos por Idosos. Latin American Journal of Pharmacy, 29, 127–31. Fisberg, R. M., Slater, B., Barros, R. R., Lima, F. D., Cesar, C. L. G., Carandina, L., et al (2004) Healthy Eating Index: evaluation of adapted version and its applicability. Revista de Nutrição, 17:301-308. Flores, L. M. & Mengue, S. S. (2005). Uso de medicamentos por idosos em região do sul do Brasil. Revista de Saude Publica, 39, 924–929. Flores, V., & Benvegnú, L. (2008). Use of medicines by the elderly in Santa Rosa, Rio Grande do Sul State, Brazil. Cadernos de Saúde Pública, 24, 1439–1446. Francisco, P. M. S. B., Segri, N. J., Barros, M. B. A., & Malta, D. C. (2015). Sociodemographic inequalities in non communicable chronic disease risk and protection factors: telephone survey in Campinas, São Paulo, Brazil. Epidemiologia & Serviços de Saúde, 24, 7-18. Gadenz, S. D., & Benvegnú, L. A. (2013). Eating habits in the prevention of cardiovascular diseases and associated factors in elderly hypertensive individuals. Ciência & Saúde Coletiva, 18, 3523–3534. Giuli, C. et al. (2012) Habits and ageing in a sample of Italian older people. The Journal of Nutrition Health and Aging, 16, 875–879. Gontijo, M. F. et al. (2012). Factors Associated with the Use of Hypoglycemic and Antihypertensive Drugs among the Elderly , Living in a South-Eastern Capital City of Brazil. Latin American Journal of Pharmacy, 31, 574-581. Guenther, P.M., DeMaio, T.J., Ingwersen, L.A., Berlin, M. (1995). The Multiple-pass Approach for the 24-hour Recall in the Continuing Survey of Food Intakes by Individuals (CSFII) 1994-96. Boston, Mass: International Conference on Dietary Assessment Methods. Guenther, P. M., Reedy, J., & Krebs-Smith, S.M. (2008 a). Development of the Healthy Eating Index-2005. Journal of the American Dietetic Association, 108, 1896–901. Guenther, P. M., Reedy, J., & Krebs-Smith, S. M. (2008 b). Evaluation of the Healthy Eating Index-2005. Journal of the American Dietetic Association, 108, 1854–1864. Hiza H. A. B., Casavale K. O., Guenther, P.M., & Davis C.A. (2013). Diet Quality of Americans Differs by Age, Sex, Race/Ethnicity, Income, and Education Level, Journal of the Academy of Nutrition and Dietetics, 113, 297–306. Jenkins, D. J. A., et al. (2008). Effect of a low-glycemic index or a high-cereal fiber diet on type 2 diabetes: a randomized trial. Journal of the American Medical Association, 300, 2742–2753. Kant, A. K. (2010) Dietary patterns: Biomarkers and chronic disease risk. Applied Physiology, Nutrition, and Metabolism, 35, 199–206.
Lima-Costa, M. F., Peixoto, S. V., & Firmo, J. O. A. ( 2004). Validade da hipertensão arterial auto-referida e seus determinantes (projeto Bambuí). Revista de Saúde Pública, 38, 637-642. Lopes, A. C. S. et al. (2005). Consumo de nutrientes em adultos e idosos em estudo de base populacional: Projeto Bambuí. Cadernos de Saúde Pública, 21, 1201–1209. Louzada, M. L. C., et al. (2012). Healthy eating index in southern brazilian older adults and its association with socioeconomic, behavioral and health characteristics. Journal of Nutrition, Health and Aging, 16, 3–7. Loyola Filho, A. I. et al. (2005). Estudo de base populacional sobre o consumo de medicamentos entre idosos: Projeto Bambuí. Cadernos de Saúde Pública, 21, 545–553. Malta, D. C., Moura, E. C., & Morais Neto, O. L. (2011). Gender and schooling inequalities in risk and protective factors for chronic diseases among Brazilian adults , through telephone survey. Revista Brasileira de Epidemiologia, 14, 125–135.
Martin, L. M., Leff, M., Calonge, N., Garrett, C., & Nelson, D. E. (2000). Validation of selfreported chronic conditions and health services in a managed care population. American Journal of Preventive Medicine, 18, 215-218. Nascimento, C. M., Ribeiro, A. Q., Cotta, R. M. M. et al (2011). Nutritional status and associated factors among the elderly in Viçosa, Minas Gerais State, Brazil. Cadernos de Saúde Pública, 27, 2409–2418. Nascimento, C. D. M. et al. (2012). Factors associated with functional ability in Brazilian elderly. Archives of gerontology and geriatrics, 54, e89–94. Núcleo de Estudos e Pesquisas em Alimentação (2011). Tabela Brasileira de Composição de Alimentos – TACO. 4ª ed. Campinas: Universidade Estadual de Campinas. http://www.unicamp.br/nepa/taco/contar/taco_versao4.pdf Accessed 10.03.2015. Pinheiro, A. B. V., Lacerda, E. M. A., Benzecry, E. H., Gomes, M. C. S., & Costa, V.M. (2005). Tabela para Avaliação de Consumo Alimentar em Medidas Caseiras. (5th ed.). Atheneu: São Paulo. Previdelli, Á. N., Andrade, S. C., Pires, M. M., Ferreira, S. R. G., Fisberg, R. M., Marchioni, D. M. (2011) A revised version of the Healthy Eating Index for the Brazilian population. Revista de Saúde Pública, 45, 794–798. Rocha, F. L. et al. (2013). Correlação entre indicadores de obesidade abdominal e lipídeos séricos em idosos. Revista da Associação Médica Brasileira, 59, 48–55. Romaguera, D. et al. (2011). Food Composition of the Diet in Relation to Changes in Waist Circumference Adjusted for Body Mass Index. PLoS ONE, 6, e23384.
Savoca, M. R. et al. (2009). The Diet Quality of Rural Older Adults in the South as Measured by HEI-2005 Varies by Ethnicity. Journal of the American Dietetic Association, 18, 1199–1216. Selem, S. S. C., Castro, M. A., César, C. L. G., Marchioni, D. M. L., & Fisberg, R. M. (2013). Validade da hipertensão autorreferida associa-se inversamente com escolaridade em brasileiros. Arquivos Brasileiros de Cardiologia, 100, 52-59. Scagliusi, F. B., & Lancha Júnior, A. H.(2003). Underreporting of energy intake in dietary assessment methods. Revista de Nutrição, 16, 471–481. Silva, N. D. A., Pedraza, D. F., & Menezes, T. N. (2015). Desempenho funcional e sua associação com variáveis antropométricas e de composição corporal em idosos. Ciência & Saúde Coletiva, 20, 3723–3732. Venturini, C. D., Engroff, P., Sgnaolin, V. et al. (2015). Consumption of nutrients among the elderly living in Porto Alegre in the State of Rio Grande do Sul, Brazil: a population-based study. Ciência & Saúde Coletiva, 20, 3701–3711. Viebig, R. F., Pastor-Valero, M., Scazufca, M., & Menezes, P. R. (2009). Fruit and vegetable intake among low income elderly in the city of São Paulo, Southeastern Brazil. Revista de Saúde Pública, 43, 806–813. World Health Organization (WHO), 1995. Physical status: the use and interpretation of anthropometry. World Health Organization technical report series, Geneva.
Table 1 – Socioeconomic characteristics of the elderly. Viçosa, Brazil, 2009. Variable
n
%
Gender Female
330
53,2
Male
290
46,8
433
69,8
187
30,2
94
15,2
Early grades of elementary school
396
64,0
Latest grades of elementary school or higher
129
20,8
72
11,8
Q2 ($ 847.88 – 952.71)
233
38,2
Q3 ($ 952.72 – 2691.78)
153
25,0
Q4 (≥ $ 2691.79)
153
25,0
66
10,6
554
89,4
Age range (years) 60-74 ≥ 75 Education Never studied
Individual income (quartiles) Q1 ($ 0 – 847.87)
Living standards Living alone Living with another
Table 2 – Characteristics regarding health, use of health services, lifestyle, waist circumference and diet quality of the elderly. Viçosa, Brazil, 2009. Variables Self-perfection of health Very good/good
n
%
272
45,4
Average
289
48,3
Bad/very bad
38
6,3
Medical appointments (previous year) None
45
7,3
1 to 5
449
72,5
≥6
125
20,2
526
85,0
93
15,0
519
84,0
99
16,0
Number of hospitalizations (previous year) None ≥1 Functional capacity Adequate Inadequate Difficulty to eat
53
8,6
Difficulty to prepare food
66
16,1
Polypharmacy (past 15 days)
224
36,1
Number of disorders ≤4
385
62,1
235
37,9
History of diabetes mellitus
138
22,3
History of dyslipidemia
352
56,9
History of high blood pressure
474
76,5
Alcohol consumption Yes
≥5
209
33,8
No but have drunk before
205
33,1
Never drank
205
33,1
Smoking Yes
67
10,8
No but have smoked before
206
33,4
Never smoked
345
55,8
78
12,6
Average
Min-Max
Reduction of food intake (past 3 months) Waist circumference (cm) Tertile 1
82,74
61.30-89,69
Tertile 2
95,18
89.70-100,79
109,21
100.80-155,80
Average
Min-Max
Tertile 1
51,72
21,95-59,66
Tertile 2
64,62
59,67-70,05
Tertile 3 BHEI-R: Brazilian Healthy Eating Index–Revised.
76,43
70,10-90,56
Tertile 3 1
Total BHEI-R (points)
1
Table 3 – Univariate analysis between the socioeconomic variables and those for diet quality *, Viçosa, Brazil, 2009. Variable
BHEI-R Best diet quality
BHEI-R Intermediate diet quality n (%)
OR (IC 95%)
n (%)
OR (IC 95%)
Gender Male
92 (44,5)
1,0
81 (39,1)
1,0
Female
115 (55,6)
1,6 (1,1-2,4)
126 (60,9)
2,0 (1,4-3,0)
Age range (years) 60-74
147 (71,0)
1,0
147 (71,0)
1,0
≥ 75
60 (29,0)
0,8 (0,6-1,3)
60 (29,0)
38 (18,4)
1,0
63 (30,4)
1,0
135 (65,2)
1,2 (0,7-2,0)
125 (60,4)
2,0 (1,1-3,6)
34 (16,4)
1,6 (0,8-3,2)
19 (9,2)
4,9 (2,4-9,8)
0,8 (0,6-1,3)
Education Never studied Early grades of elementary school Latest grades of elementary or higher Individual income (quartiles) Q4 (≥ R$ 1476.25)
51 (24,8)
1,0
61 (30,2)
1,0
Q3 (R$ 522.50 – 1476.24)
54 (26,2)
0,9 (0,5-1,7)
50 (24,8)
0,6 (0,3-1,2)
Q2 (R$ 465.00 – 522.49)
78 (37,9)
1,1 (0,5-2,2)
65 (32,2)
0,9 (0,5-1,8)
Q1 (R$ 0 – 464.99)
23 (11.2) 1,3 (0,7-2,6) 26 (12,9) 1,3 (0,7-2,6) IC95%: 95% confidence interval; OR: odds ratio; PC: waist circumference. BHEI-R: Brazilian Healthy Eating Index–Revised. *For this analysis, the reference category was the lowest for the tertile with BHEI-R of “Poor diet quality”.
Table 4 - Univariate analysis of the association between the variables health, use of health services, lifestyle, waist circumference and diet quality*, Viçosa, Brazil, 2009. Variable
BHEI-R Intermediate diet quality n (%) OR (IC 95%)
BHEI-R Best diet quality n (%) OR (IC 95%)
Functional capability Adequate
174 (84,1)
1,0
169 (82,4)
1,0
Inadequate
33 (16,0)
1,1 (0,6-1,9)
36 (17,6)
1,2 (0,7-2,1)
Difficulty to eat No
189 (91,3)
1,0
191 (92,3)
1,0
18 (8,7)
0,9 (0,5-1,8)
16 (7,7)
0,8 (0,4-1,7)
Difficulty to prepare food No
126 (86,3)
1,0
112 (81,8)
1,0
Yes
20 (13,7)
0,8 (0,4-1,6)
25 (18,2)
1,1 (0,6-2,2)
Medical appointments (past year) None
Yes
17 (8,2)
1,0
5 (2,4)
1,0
1 to 5
148 (71,5)
1,4 (0,7-2,7)
156 (75,7)
4,9 (1,8-13,4)
≥6
42 (20,3)
1,5 (0,7-3,2)
45 (21,8)
5,4 (1,9-15,7)
Number of disorders ≤4
138 (66,7)
1.0
104 (50,2)
1,0
≥5
69 (33,4)
1,1 (0,8-1,7)
103 (49,8)
2,2 (1,5-3,4)
History of diabetes mellitus No
169 (81,6)
1,0
131 (63,3)
1,0
Yes
38 (18,4)
1,7 (1,0-3,0)
76 (36,7)
4,4 (2,6-7,3)
History of dyslipidemia No
96 (46,4)
1,0
62 (29,9)
1,0
Yes
110 (53,4)
1,3 (0,9-1,9)
145 (70,0)
2,6 (1,8-3,9)
History of high blood pressure No
54 (26,1)
1,0
39 (18,8)
1,0
Yes
153 (73,9)
1,0 (0,6-1,5)
168 (81,2)
1,5 (0,9-2,4)
Polypharmacy No
140 (67,6)
1,0
104 (50,2)
1,0
Yes
67 (32,4)
1,3 (0,9-2,1)
103 (49,8)
2,8 (1,8-4,2)
Alcohol consumption Never drank
73 (35,3)
1,0
69 (33,5)
1,0
No but has drunk before
65 (31,4)
0,8 (0,5-1,4)
70 (34,0)
1,0 (0,6-1,6)
Yes
69 (33,4)
0,9 (0,6-1,5)
67 (32,5)
0,9 (0,6-1,5)
Smoking Never smoked
115 (55,8)
1,0
122 (59,2)
1,0
No but has smoked before
73 (35,4)
1,0 (0,6-1,5)
62 (30,1)
0,8 (0,5-1,2)
Yes
18 (8,7)
0,6 (0,3-1,2)
22 (10,7)
0,7 (0,4-1,3)
Reduction of food intake No
184 (88,9)
1,0
182 (87,9)
1,0
Yes
23 (11,1)
0,7 (0,4-1,3)
25 (12,1)
0,8 (0,5-1,4)
PC (tertiles)
Lower than the first tertile
54 (26,7)
1,0
65 (32,2)
1,0
Between the first and second tertiles
80 (40,8)
2,4 (1,5-3,9)
65 (33,2)
1,6 (1,0-2,7)
Higher our equal to second tertile
66 (33,3) 1,6 (0,9-2,5) 67 (33,8) 1,3 (0,8-2,1) IC95%: 95% confidence interval; OR: odds ratio; PC: waist circumference. BHEI-R: Brazilian Healthy Eating Index–Revised. *For this analysis, the reference category was the lowest for the tertile with BHEI-R of “Poor diet quality”.
Table 5 - Results of the multivariate analysis of the association between sociodemographic conditions, health status, use of health services, lifestyle, waist circumference and diet quality *, Viçosa, Brazil, 2009. BHEI-R Intermediate diet quality OR (IC 95%)
BHEI-R Best diet quality OR (IC 95%)
Gender Male
1,0
1,0
Female
1,8 (1,1-2,9)
1,7 (1,0-2,9)
Variable
Education Never studied
1,0
1,0
Early grades of elementary school
1,4 (0,8-2,4)
3,1 (1,5-6,1)
Latest grades of elementary or higher
2,0 (0,9-4,1)
10,5 (4,5-24,5)
Medical appointments (past year) None
1,0
1,0
1 to 5
1,2 (0,6-2,6)
≥6
1,2 (0,5-2,9)
3,7 (1,0-13,1) 3,2 (0,8-12,5)
1,0
1,0
1,4 (0,8-2,6)
4,1 (2,3-7,5)
1,0
1,0
1,1 (0,7-1,7)
1,6 (1,0-2,6)
1,0
1,0
0,8 (0,5-1,3)
1,1 (0,7-2,0)
1,0
1,0
1,1 (0,7-1,8)
1,8 (1,1-3,0)
History of diabetes mellitus No Yes History of dyslipidemia No Yes History of high blood pressure No Yes Polypharmacy No Yes Alcohol consumption Never drank
1,0
1,0
No but has drunk before
1,3 (0,8-2,0)
0,8 (0,5-1,4)
Yes
0,9 (0,4-1,9)
1,4 (0,7-2,9)
1,0
1,0
2,5 (1,5-4,2)
1,6 (0,9-2,7)
PC (tertiles) Lower than the first tertile Between the first and second tertiles Higher our equal to second tertile
1,5 (0,9-2,4) 0,9 (0,5-1,5) IC95%: 95% confidence interval; OR: odds ratio; PC: waist circumference. BHEI-R: Brazilian Healthy Eating Index–Revised. *For this analysis, the reference category was the lowest for the tertile with BHEI-R of “Poor diet quality”.
S0167-4943(17)30230-3 http://dx.doi.org/doi:10.1016/j.archger.2017.05.006 AGG 3482
To appear in:
Archives of Gerontology and Geriatrics
Received date: Revised date: Accepted date:
17-10-2016 6-5-2017 9-5-2017
Please cite this article as: Pinto de Souza Fernandes, Dalila, Duarte, Maria Sˆonia Lopes, Pessoa, Milene Cristine, Franceschini, Sylvia do Carmo Castro, Ribeiro, Andr´eia Queiroz, EVALUATION OF DIET QUALITY OF THE ELDERLY AND ASSOCIATED FACTORS.Archives of Gerontology and Geriatrics http://dx.doi.org/10.1016/j.archger.2017.05.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
EVALUATION OF DIET QUALITY OF THE ELDERLY AND ASSOCIATED FACTORS Dalila Pinto de Souza Fernandes1 Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil (32) 98428-5058 [email protected] Maria Sônia Lopes Duarte1 Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil (31) 3899-3741 [email protected] Milene Cristine Pessoa2 Department of Nutrition, Nursing school, Federal University of Minas Gerais, Av. Prof. Alfredo Balena, 190, Santa Efigênia, Belo Horizonte, MG, 30130-100, Brazil (31) 3409-9179 [email protected] Sylvia do Carmo Castro Franceschini1 Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil (31) 3899-3743 [email protected] Andréia Queiroz Ribeiro1 Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil (31) 3899-1271 [email protected]
1
2
Federal University of Viçosa Federal University of Minas Gerais
Please address all correspondence to: Dalila Pinto de Souza Fernandes Department of Nutrition, Federal University of Viçosa, Av. PH Rolfs, s/n., Campus, Viçosa, MG, 36570-000, Brazil e-mail: [email protected] Tel: +55 – 32 – 98428-5058
ABSTRACT Background: Observational studies suggest healthy dietary patterns are associated with risk reduction and better control of various chronic diseases. However, few Brazilian studies have focused on evaluating the quality of the elderly diet and its relationship with diseases. This study aimed to estimate the association between diet quality and socioeconomic factors, health and nutrition of the elderly. Methods: This is a cross-sectional population-based study whose target population were non-institutionalized elderly residents in the city of Viçosa, Brazil. Anthropometric, socioeconomic, health conditions, lifestyle and food consumption variables were obtained from a semi-structured questionnaire. The quality of the diet was assessed by the revised Healthy Eating Index classified into tertiles, considering the first tertile as "Poor diet quality," the second as "Intermediate diet quality" and the third as "Better diet quality." To identify factors independently associated with diet quality model, the works used multinomial logistic regression. Results: In the results of the multivariate analysis, the factors independently associated with "better diet quality" included female gender, higher education, history of one to five medical visits in the past year, history of diabetes mellitus, dyslipidemia and the use of polypharmacy. Discussion: Our results show that most seniors need to improve the quality of their diet and those of male gender with no or little education, and those who do not seek medical services constitute the group that needs attention concerning the measures to improve the quality of their diet.
Keywords: Diet; Diet quality; Food consumption; Elderly; Elderly health.
Highlights The diet quality of brazilian elderly requires improvement. Elderly women and those with some education level presented better diet quality. Elderly with diabetes mellitus and dyslipidemia presented better diet quality.
Elderly who do not seek health services are the groups that needs more attention.
EVALUATION OF DIET QUALITY OF THE ELDERLY AND ASSOCIATED FACTORS 1. Introduction
Eating habits play a unique role in health, as the combination of food consumed over time throughout the life cycle can be determinant of the occurrence of various diseases (Hiza et al., 2013). Results from observational studies suggest that healthy dietary patterns are associated with risk reduction and better control of several chronic diseases such as obesity, dyslipidemia, hypertension and diabetes mellitus (Lopes et al., 2005; Kant, 2010; Giuli et al., 2012; Hiza et al., 2013). Among the elderly, nutrition becomes particularly essential for the good maintenance of acceptable health standards and functional capacity (Silva, Pedraza & Menezes, 2015). In order to properly estimate the relationship between diet and morbidities, specific tools have been developed to assess food consumption (Fisberg et al., 2004). Because of the relationship between food, the variety of factors associated with its intake and the prevalence of certain diseases that increase with advancing age, we consider of great importance to conduct the diet evaluation globally since it would better reflect the complexity of this relationship (Louzada et al., 2012). In this context, the Brazilian Healthy Eating Index–Revised (BHEI-R) (Previdelli et al., 2011)
was
proposed as dietary assessment tool, wich is based on the food guide recommendations for the Brazilian population (BRASIL, 2008) and the Healthy Eating Index of 2005 (Guenther, Reedy & Krebs-Smith, 2008a). Few Brazilian studies published in scientific literature have focused on evaluating the diet quality of the elderly. Among the published studies, the authors have observed an association between a poorer diet quality and low education, low income, lack of appetite, difficulty in purchasing and preparing food (Viebig et al., 2009) in addition to smoking, consumption of sugary drinks and alcohol (Assumpção et al., 2014). On the other hand, the elderly with nutritional counseling, non-smokers and those who practice physical activity have shown better diet quality (Gadenz, Benvegnú, 2013; Assumpção et al., 2014). However, there are few studies with Brazilian elderly that assessed dietary intake through the BHEI-R and that took chronic diseases that may be related to food into consideration. In this sense, the present study aimed to estimate
the association between diet quality and specific socioeconomic factors, health and nutrition of the elderly.
2. Methodology 2.1. Designing the study This is a cross-sectional population-based study, which integrates the research project "Health conditions, nutrition and medication use by the elderly in Viçosa (MG): a population-based survey" conducted from June to December 2009. This project was approved by the Ethics Committee on Human Research of the Federal University of Viçosa (Official letter No. 27/2008/CEP/UFV). The evaluation of the elderly was carried out only after all parties signed the Free and Informed Consent Term (FICT).
2.2. Target population and sample The study’s target population consisted of seniors aged over 60 years, who are not institutionalized, residents in rural and urban areas of Viçosa, which corresponded to 7980 elderly in 2008 (study period). Those data were the basis for the calculation of the sample. The sample size was calculated considering a confidence level of 95%, estimated prevalence of 50% for different outcomes of interest to the larger project, 4% tolerated error and 20% increase to cover losses. From an initial study sample of 670 participants, 50 were excluded because there were losses due to subject refusal to participate (3.6%), unavoidable reasons that interviews could not be conducted such as in situations where randomly selected individuals had died (1.3%), had an address that could not be located in the city (1.2%) or had moved to other locations which were either difficult to find or in other municipalities (1,2%), in addition to missing data from the recall of habitual consumption (0,15%). Those participants excluded were no different from the included ones regarding gender and age range (Nascimento et al., 2012). Thus, excluding losses, the final sample was of 620 elderly, what represents a response rate of 93.0%.
2.3. Data collection
The interviews were previously scheduled, they were performed at each elderly’s home and conducted by pairs of trained interviewers. Up to three attempts to visit the interviewee’s home were performed before the person was considered a loss. The data collection instrument consisted of a semi-structured questionnaire with sociodemographic variables, health conditions, lifestyle and food consumption. To obtain the food consumption data we have applied a standard recall of habitual consumption using the 'multi-pass' technique (Guenther, P.M., et al., 1995). The contents of the recall were reviewed prior to data entry. Therefore, we have performed a standardization of preparations and percentage of salt and oil (Pinheiro et al., 2005; Araújo, Guerra, 2007; TACO, 2011). Added sugars included all sugars added to foods during processing, preparation or consumption. The amount of added sugar in each food was standardized based on the Table for Evaluation of Food Consumption in Household Measurements (Pinheiro et al., 2005), in recipes and on foods labels. Later, the data were entered in Dietpro® version 5i software, and it were performed quality control of the typed data. The study has also measured the interviewee’s waist circumference (WHO, 1995).
2.4. Study variables The Brazilian Healthy Eating Index–Revised (BHEI-R), which evaluates the diet quality and was obtained from the information from the recall of habitual consumption, was used as dependent variable. This index was revised by Previdelli et al (2011) and validated for Brazilian population by Andrade et al (2013). In relation to the total score, the maximum value of the total BHEI-R is 100 points. Values closer to the maximum score represent better diet quality. Because the index is designed to reflect different aspects of the diet, there is no classification in adequate and inadequate considering the total score (Guenther, Reedy & Krebs-Smith, 2008a; Guenther, Reedy & Krebs-Smith, 2008b). Therefore, the total BHEI-R was ranked into tertiles, considering the first tertile as "Poor diet quality," the second as "Intermediate diet quality" and the third as "Best diet quality." The independent variables in this study included: Socioeconomic: gender (male, female); age range (60-74 years, 75 years and older); education (never studied; early grades of elementary school; final grades of elementary school or higher); individual income (quartiles) and cohabitation (living alone, living together).
Health conditions: number of medical appointments (none; 1-5; 6 or more) during the year before the interview and number of hospitalizations (none, 1 or more) during the year before the interview. Functional capacity was assessed by questions about ability to perform 14 basic and instrumental activities of daily living, categorized into: no struggle, little struggle, great struggle, unable to do it and does not do it (Nascimento et al., 2012). In the analysis, we have considered the variables difficulty to feed and difficult to prepare food, as well as functional capacity defined as inadequate, when the elderly reported difficulty performing seven or more activities, or when self-assessed as unable to perform three or more activities; and otherwise as appropriate. We have also evaluated the polypharmacy (yes, no), defined as the use of five or more drugs over the 15 days preceding the interview2 (Beloosesky et al., 2013). As for morbidities, information about the history of chronic diseases were obtained from the following question: "Have you ever in life had a doctor or other health professional said that you had or have had any of the following diseases?". Diseases of interest to the study included diabetes mellitus; arthrosis, arthritis or rheumatism; heart attack; asthma or bronchitis; depression; hearing problems; high pressure; angina; stroke; vision problems; osteoporosis; dyslipidemia; kidney disease and cancer. For the analysis we used the total number of disorders and history of diabetes mellitus, dyslipidemia and high blood pressure.
Lifestyle: alcohol consumption (yes; no but has drunk before; never drank) and smoking (yes; no but has smoked before; never smoked).
Anthropometric measure: we have measured the waist circumference (WC) using flexible and inelastic tape (TBW®, São Paulo, Brazil) 1.80m long and 0.1mm accuracy. This variable was measured with the elderly in the standing position, the height of the midpoint between the last rib and the iliac crest, during expiration, according to the protocol of the World Health Organization, 1995 (WHO, 1995).
2.5. Data analysis For this study, we proceeded the descriptive analysis of variables of interest, based on the frequency distribution and estimates of central tendency and dispersion. The normal distribution of the BHEI-R values between the categories of the variables was assessed using the Shapiro-Wilk test.
Multinomial logistic regression analysis was used to obtain estimates for Odds Ratios and 95% confidence intervals of the association between the independent variables of interest in the study and the diet quality. This method of regression is used when the dependent variable has more than two categories, each being compared to a reference category. In this study the reference category was the lowest tertile values of BHEI-R, which was called "Poor diet quality." The variables that were associated with the dependent variable in univariate step with p<0.20 were included in the multivariate multinomial logistic regression model. The variables that were associated with the dependent variable at the level of p<0.05 remained in the final model. All analyzes were performed using STATA software (Stata Corp. College Station, United States) version 13.0, adopting α = 0.05 as statistical significance level for all comparisons.
3. Results
The sample consisted of 620 elderlies. According to the socioeconomic characteristics, most were female (53.2%), aged between 60 and 74 years (69.8%), studied until the early grades of elementary school (64.0%) and did not live alone (89.4%). About 38.0% of the elderly had individual monthly income between $ 847.88 and $ 952.71 (second quartile), whereas the lower limit corresponds to the minimum wage during the period of the study (Table 1). Regarding to health and food consumption, the sample characterization is shown in Table 2. We have identified that 48.3% of the elderly rated their health as fair, most (72.5%) had a history of one to five doctor appointments and 15.0% had a hospital history during the previous year. Functional capacity was classified as inappropriate in 16.0% of the elderly, and 8.6% reported difficulty to feed and 16.1% reported difficulty to prepare food. Polypharmacy was observed in 36.1% of seniors and 38,0% reported having five or more diseases. About 22.0% had a history of diabetes mellitus, 56.0% of dyslipidemia and 76.5% of high pressure. In relation to alcohol consumption, 33.8% admitted to drinking and just over half of the sample reported never having smoked. Approximately 13,0% of the sample reported some reduction in food intake over the three months prior to the survey. Waist circumference and total BHEI-R are shown in tertiles (Table 2). The results of the univariate analysis showed that the factors positively
associated with the tertile "Intermediate diet quality" included female gender, having history of diabetes mellitus and having waist circumference between 89,70 and 100,79 cm. For the tertile "Best diet quality," the positive and significant associations were female gender, having studied at least until early grades of elementary school, having made at least one medical visit in the past year, having history of five or more diseases, polypharmacy, history of diabetes mellitus and dyslipidemia, and having waist circumference between 59,67 and 70,05cm (Tables 3 and 4). The results of multivariate analysis allowed us to observe the factors which remained independently associated to the tertile "Intermediate diet quality" included female gender and having waist circumference between 89,70 and 100,70cm. On the other hand, the factors independently associated with tertile of "Best diet quality" were female gender, having higher education, having had five medical appointments in the past year, having history of diabetes mellitus, dyslipidemia and having practiced polypharmacy. The variable for number of disorders was not included in this multivariate analysis because the variables history of diabetes mellitus and history of dyslipidemia were included in the first (Table 5).
4. Discussion
The results indicate that the diet quality of the elderly requires improvement, considering that two-thirds had scores ranging between about 22 and 70 points. This finding is also reported by other authors (Ervin, Dye, 2008; Savoca et al., 2009; Cannella, Savina & Donini, 2009; Louzada et al., 2012; Assumpção et al., 2014). In this study, the goal was to investigate the association of socioeconomic variables, health status, use of health services, lifestyle and waist circumference with the elderlies’ tertiles for BHEI-R. Being female remained positively associated with tertiles "Intermediate diet quality" and "Best diet quality". Among the elderly, 55.6% belonged to the tertile of "Intermediate diet quality" and 60.9% to tertile "Best diet quality". International and national studies show that women generally have better diet quality than men(Louzada et al., 2012; Malta, Moura & Morais Neto, 2011; Hiza et al., 2013), which has been justified by the fact that they have a greater concern for their health, and with healthier eating habits (Francisco et al., 2015). Having education was positively associated with "Best diet quality", i.e. the elderly who have a healthier diet had a greater chance of having education compared
those in tertile "Poor diet quality". The strength of association was higher the higher the level of education. This association was also found in other studies (Fisberg et al., 2004; Ervin, Dye, 2008; Basiotis et al., 2002). Elderly American participants from the National Health and Nutrition Examination Survey 2003-2004 in which higher levels of education also showed better total score of HEI (Hiza et al., 2013). In a study with elderly in the Brazilian South, it was observed that those with higher education consume less carbohydrates and more foods rich in minerals and vitamins compared to those with less than 8 years of schooling, noting that the level of education is crucial to healthy food habits (Venturini et al., 2015). When the elderly showed one to five medical appointments during the year before the interview, this was positively associated with a better diet quality, as in a study with elderly in Southern Brazil (Gadenz, Benvegnú, 2013). There was also better diet quality among older people who reported a history of diabetes mellitus, dyslipidemia and use of polypharmacy. All these findings can be attributed to what is called reverse causality, as in this case, the diagnosis of a disease may raise a need for change in lifestyle in the individual, and may thus result in the adoption of healthier eating habits. The positive association between history of diabetes mellitus and "Best diet quality" probably results from the change in behavior, since diet is one of the key factors for glycemic control (Jenkins et al., 2008; Exebio et al., 2011). In a populationbased study for elderly people in São Paulo, we found similar results with better diet quality among those diagnosed with diabetes mellitus (Assumpção et al., 2014).
In
addition to diabetes mellitus, dyslipidemia and obesity are also common in this age group33, and the appropriate control and prevention of complications from these diseases are associated with healthy eating habits (Kant, 2010; Giuli et al., 2012). Our results, although consistent with the literature, have the limitation of not assessing whether the elderly with better diet quality had adequate control of the disease because no information was available for this evaluation. This aspect can be explored in future studies, especially those of longitudinal design to allow more evidence about the influence of the diet quality over the control of chronic non-communicable diseases in the elderly. Polypharmacy appears as a frequent practice among the elderly and its prevalence in the sample (36.1%) was higher than that observed in other Brazilian studies (5-27%) (Flores, Mengue, 2005; Loyola Filho et al., 2005; Flores, Benvegnú,
2008; Duarte et al., 2012; Gontijo et al., 2012) On the one hand, polypharmacy can be seen as an indicator of increased morbidity among the elderly, compared to its strong association with other diseases. In this study, its association with the "Best diet quality" may indicate that seniors using five or more medications are more concerned with healthy eating. Notwithstanding, there is an important concern about the influence of using multiple drugs in the elderly nutritional status. Certain studies have evaluated its association with nutritional factors and it is known that the use of multiple drugs, inappropriate prescribing and use of two or more drugs with the same pharmacological activity, among other factors, promote increased side effects and interactions that interfere with digestion, absorption and metabolism of nutrients (Faria, Franceschini & Ribeiro, 2010).
Additional studies are needed to verify the actual interference of
polypharmacy in the bioavailability of nutrients as well as longitudinal studies can help to elucidate the influence of polypharmacy in the nutritional status of the elderly. As the high concentration of serum lipids, abdominal obesity is a major risk factor for cardiovascular diseases (Rocha et al., 2013). As waist circumference (WC) is an indicator of central adiposity, it has been associated with the risk of morbidity and mortality in the elderly (Romaguera et al., 2011). In the present study, only the second tertile for WC values remained associated with "Intermediate quality diet", which also suggests a change in behavior. The fact that the WC tertile with the highest values is not associated with the "Best diet quality" can mean low sample power, given that the study was not specifically designed to evaluate this relationship. Some limitations should be considered when interpreting the findings of the present study. First, its cross-sectional design limits the interpretation of some associations found as a result of cause and effect relationship. However, the results are consistent with the literature. Second, some variables analized in this study were selfreported, however, studies in adults and the elderly attest the reliability of this information (Martin, et al., 2000; Lima-Costa, Peixoto & Firmo, 2004; Selem, et al., 2013). Third, the variable physical activity was not considered as a control, due to the lack of information about kind of activity, intensity and frequency. At this point of view, the relationship between physical activity practice and diet quality improvement should be adressed in future studies. Fourth, there are the limitations inherent to dietary surveys such as errors in the estimate of food portions and memory bias (Scagliusi, Lancha Júnior, 2003). One reservation should be made to the fact that we have used the recall of habitual consumption as consumer assessment tool, which despite not being
validated, allowed us to observe that the elderly tend to have monotonous diets as the difficulties to prepare food and even to eat increase (Venturini et al., 2015) and even due to income issues, which minimizes the potential bias in this case and justifies the use of this recall. Moreover, this recall provides a better estimate of food intake when compared to the Food Frequency Questionnaire, which has a limited number of food items. This study’s results show most seniors need to improve their diet quality and there is evidence of the association between female gender, having higher education, having history of diabetes mellitus, dyslipidemia, using polypharmacy and having the "Best quality diet". Considering some methodological differences between the studies, these results are similar to those observed in the elderly in developed countries and larger cities in Brazil. Elderly men, with any or little education and those who do not seek health services are the groups that needs more attention regarding to the measures to improve their diet quality.
Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest. Authors’ contributions Fernandes DPS worked in study design, analysis and interpretation of data, article writing and critical review of relevant intellectual content and final approval of the version to be published. Ribeiro AQ guided the design, collection and analysis and interpretation of data, and critical review of relevant intellectual content and final approval of the version to be published. Duarte MSL and Pessoa MC collaborated in the analysis and interpretation of data, article writing and critical review of relevant intellectual content; Franceschini SCC collaborated in the relevant critical review of the intellectual content.
Ethical The study was conducted according to the directives established in the Declaration of Helsinki, and all procedures involving human being were approved by the Research Ethics Committee of the Federal University of Viçosa, Brazil. All persons gave their informed consent prior to their inclusion in the study.
Acknowledgments The Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), (process n° 579255/2008-5) funded this project and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) provided Master’s degree scholarship. The authors thank the team of researchers who collected the data and the Grupo de Estudo sobre Envelhecimento, Nutrição e Saúde (GREENS) of the Department of Nutrition and Health of the Federal University of Viçosa.
References Andrade, S. C., Previdelli, Á. N., Marchioni, D. M. L., & Fisberg, R. M. (2013). Evaluation of the reliability and validity of the Brazilian Healthy Eating Index Revised. Revista de Saúde Pública, 47, 675-83. Araújo, M. O. D., & Guerra, T. M. M. (2007). Alimentos “Per capita”. (3th ed.). UFRN: Natal. Assumpção, D., Domene, S. M. Á., Fisberg, R. M., & Barros, M. B. A. (2014). Diet quality and associated factors among the elderly: a population-based study in Campinas, São Paulo State, Brazil. Cadernos de Saúde Pública, 30, 1680-1694. Basiotis, P. S. et al. The Healthy Eating Index: 1999-2000. (2002). U.S. Department of Agriculture, Center for Nutrition Policy and Promotion. CNPP-12. p. 1999–2000. Beloosesky, Y. et al. (2013). Rates, variability, and associated factors of polypharmacy in nursing home patients. Clinical Interventions in Aging, 8, 1585–1590. Brasil (2008) Guia alimentar para a população brasileira: promovendo à alimentação saudável. Secretaria de Atenção à Saúde, Coordenação-Geral da Política de Alimentação e Nutrição. Ministério da Saúde, Brasília. Cannella, C., Savina, C., & Donini, L. M. (2009). Nutrition, longevity and behavior. Archives of gerontology and geriatrics, 49, 19–27. Duarte, L. R. et al. (2012). Habits of medication consumption among elderly SUS and health insurance users. Cadernos de Saúde Coletiva, 20, 64–71. Ervin, R. B., & Dye, B. A. (2008). Eating Index scores among adults, 60 years of age and over, by sociodemographic and health characteristics: United States, 1999–2002. Advanced Data, 20, 1–16. Exebio, J. C. et al. (2011). Healthy Eating Index scores associated with symptoms of depression in Cuban-Americans with and without type 2 diabetes: a cross sectional study. Nutrition Journal, 10, 135-141.
Faria, M. Q., Franceschini, S. C. C., & Ribeiro, A. Q. (2010). Estado Nutricional e Uso de Medicamentos por Idosos. Latin American Journal of Pharmacy, 29, 127–31. Fisberg, R. M., Slater, B., Barros, R. R., Lima, F. D., Cesar, C. L. G., Carandina, L., et al (2004) Healthy Eating Index: evaluation of adapted version and its applicability. Revista de Nutrição, 17:301-308. Flores, L. M. & Mengue, S. S. (2005). Uso de medicamentos por idosos em região do sul do Brasil. Revista de Saude Publica, 39, 924–929. Flores, V., & Benvegnú, L. (2008). Use of medicines by the elderly in Santa Rosa, Rio Grande do Sul State, Brazil. Cadernos de Saúde Pública, 24, 1439–1446. Francisco, P. M. S. B., Segri, N. J., Barros, M. B. A., & Malta, D. C. (2015). Sociodemographic inequalities in non communicable chronic disease risk and protection factors: telephone survey in Campinas, São Paulo, Brazil. Epidemiologia & Serviços de Saúde, 24, 7-18. Gadenz, S. D., & Benvegnú, L. A. (2013). Eating habits in the prevention of cardiovascular diseases and associated factors in elderly hypertensive individuals. Ciência & Saúde Coletiva, 18, 3523–3534. Giuli, C. et al. (2012) Habits and ageing in a sample of Italian older people. The Journal of Nutrition Health and Aging, 16, 875–879. Gontijo, M. F. et al. (2012). Factors Associated with the Use of Hypoglycemic and Antihypertensive Drugs among the Elderly , Living in a South-Eastern Capital City of Brazil. Latin American Journal of Pharmacy, 31, 574-581. Guenther, P.M., DeMaio, T.J., Ingwersen, L.A., Berlin, M. (1995). The Multiple-pass Approach for the 24-hour Recall in the Continuing Survey of Food Intakes by Individuals (CSFII) 1994-96. Boston, Mass: International Conference on Dietary Assessment Methods. Guenther, P. M., Reedy, J., & Krebs-Smith, S.M. (2008 a). Development of the Healthy Eating Index-2005. Journal of the American Dietetic Association, 108, 1896–901. Guenther, P. M., Reedy, J., & Krebs-Smith, S. M. (2008 b). Evaluation of the Healthy Eating Index-2005. Journal of the American Dietetic Association, 108, 1854–1864. Hiza H. A. B., Casavale K. O., Guenther, P.M., & Davis C.A. (2013). Diet Quality of Americans Differs by Age, Sex, Race/Ethnicity, Income, and Education Level, Journal of the Academy of Nutrition and Dietetics, 113, 297–306. Jenkins, D. J. A., et al. (2008). Effect of a low-glycemic index or a high-cereal fiber diet on type 2 diabetes: a randomized trial. Journal of the American Medical Association, 300, 2742–2753. Kant, A. K. (2010) Dietary patterns: Biomarkers and chronic disease risk. Applied Physiology, Nutrition, and Metabolism, 35, 199–206.
Lima-Costa, M. F., Peixoto, S. V., & Firmo, J. O. A. ( 2004). Validade da hipertensão arterial auto-referida e seus determinantes (projeto Bambuí). Revista de Saúde Pública, 38, 637-642. Lopes, A. C. S. et al. (2005). Consumo de nutrientes em adultos e idosos em estudo de base populacional: Projeto Bambuí. Cadernos de Saúde Pública, 21, 1201–1209. Louzada, M. L. C., et al. (2012). Healthy eating index in southern brazilian older adults and its association with socioeconomic, behavioral and health characteristics. Journal of Nutrition, Health and Aging, 16, 3–7. Loyola Filho, A. I. et al. (2005). Estudo de base populacional sobre o consumo de medicamentos entre idosos: Projeto Bambuí. Cadernos de Saúde Pública, 21, 545–553. Malta, D. C., Moura, E. C., & Morais Neto, O. L. (2011). Gender and schooling inequalities in risk and protective factors for chronic diseases among Brazilian adults , through telephone survey. Revista Brasileira de Epidemiologia, 14, 125–135.
Martin, L. M., Leff, M., Calonge, N., Garrett, C., & Nelson, D. E. (2000). Validation of selfreported chronic conditions and health services in a managed care population. American Journal of Preventive Medicine, 18, 215-218. Nascimento, C. M., Ribeiro, A. Q., Cotta, R. M. M. et al (2011). Nutritional status and associated factors among the elderly in Viçosa, Minas Gerais State, Brazil. Cadernos de Saúde Pública, 27, 2409–2418. Nascimento, C. D. M. et al. (2012). Factors associated with functional ability in Brazilian elderly. Archives of gerontology and geriatrics, 54, e89–94. Núcleo de Estudos e Pesquisas em Alimentação (2011). Tabela Brasileira de Composição de Alimentos – TACO. 4ª ed. Campinas: Universidade Estadual de Campinas. http://www.unicamp.br/nepa/taco/contar/taco_versao4.pdf Accessed 10.03.2015. Pinheiro, A. B. V., Lacerda, E. M. A., Benzecry, E. H., Gomes, M. C. S., & Costa, V.M. (2005). Tabela para Avaliação de Consumo Alimentar em Medidas Caseiras. (5th ed.). Atheneu: São Paulo. Previdelli, Á. N., Andrade, S. C., Pires, M. M., Ferreira, S. R. G., Fisberg, R. M., Marchioni, D. M. (2011) A revised version of the Healthy Eating Index for the Brazilian population. Revista de Saúde Pública, 45, 794–798. Rocha, F. L. et al. (2013). Correlação entre indicadores de obesidade abdominal e lipídeos séricos em idosos. Revista da Associação Médica Brasileira, 59, 48–55. Romaguera, D. et al. (2011). Food Composition of the Diet in Relation to Changes in Waist Circumference Adjusted for Body Mass Index. PLoS ONE, 6, e23384.
Savoca, M. R. et al. (2009). The Diet Quality of Rural Older Adults in the South as Measured by HEI-2005 Varies by Ethnicity. Journal of the American Dietetic Association, 18, 1199–1216. Selem, S. S. C., Castro, M. A., César, C. L. G., Marchioni, D. M. L., & Fisberg, R. M. (2013). Validade da hipertensão autorreferida associa-se inversamente com escolaridade em brasileiros. Arquivos Brasileiros de Cardiologia, 100, 52-59. Scagliusi, F. B., & Lancha Júnior, A. H.(2003). Underreporting of energy intake in dietary assessment methods. Revista de Nutrição, 16, 471–481. Silva, N. D. A., Pedraza, D. F., & Menezes, T. N. (2015). Desempenho funcional e sua associação com variáveis antropométricas e de composição corporal em idosos. Ciência & Saúde Coletiva, 20, 3723–3732. Venturini, C. D., Engroff, P., Sgnaolin, V. et al. (2015). Consumption of nutrients among the elderly living in Porto Alegre in the State of Rio Grande do Sul, Brazil: a population-based study. Ciência & Saúde Coletiva, 20, 3701–3711. Viebig, R. F., Pastor-Valero, M., Scazufca, M., & Menezes, P. R. (2009). Fruit and vegetable intake among low income elderly in the city of São Paulo, Southeastern Brazil. Revista de Saúde Pública, 43, 806–813. World Health Organization (WHO), 1995. Physical status: the use and interpretation of anthropometry. World Health Organization technical report series, Geneva.
Table 1 – Socioeconomic characteristics of the elderly. Viçosa, Brazil, 2009. Variable
n
%
Gender Female
330
53,2
Male
290
46,8
433
69,8
187
30,2
94
15,2
Early grades of elementary school
396
64,0
Latest grades of elementary school or higher
129
20,8
72
11,8
Q2 ($ 847.88 – 952.71)
233
38,2
Q3 ($ 952.72 – 2691.78)
153
25,0
Q4 (≥ $ 2691.79)
153
25,0
66
10,6
554
89,4
Age range (years) 60-74 ≥ 75 Education Never studied
Individual income (quartiles) Q1 ($ 0 – 847.87)
Living standards Living alone Living with another
Table 2 – Characteristics regarding health, use of health services, lifestyle, waist circumference and diet quality of the elderly. Viçosa, Brazil, 2009. Variables Self-perfection of health Very good/good
n
%
272
45,4
Average
289
48,3
Bad/very bad
38
6,3
Medical appointments (previous year) None
45
7,3
1 to 5
449
72,5
≥6
125
20,2
526
85,0
93
15,0
519
84,0
99
16,0
Number of hospitalizations (previous year) None ≥1 Functional capacity Adequate Inadequate Difficulty to eat
53
8,6
Difficulty to prepare food
66
16,1
Polypharmacy (past 15 days)
224
36,1
Number of disorders ≤4
385
62,1
235
37,9
History of diabetes mellitus
138
22,3
History of dyslipidemia
352
56,9
History of high blood pressure
474
76,5
Alcohol consumption Yes
≥5
209
33,8
No but have drunk before
205
33,1
Never drank
205
33,1
Smoking Yes
67
10,8
No but have smoked before
206
33,4
Never smoked
345
55,8
78
12,6
Average
Min-Max
Reduction of food intake (past 3 months) Waist circumference (cm) Tertile 1
82,74
61.30-89,69
Tertile 2
95,18
89.70-100,79
109,21
100.80-155,80
Average
Min-Max
Tertile 1
51,72
21,95-59,66
Tertile 2
64,62
59,67-70,05
Tertile 3 BHEI-R: Brazilian Healthy Eating Index–Revised.
76,43
70,10-90,56
Tertile 3 1
Total BHEI-R (points)
1
Table 3 – Univariate analysis between the socioeconomic variables and those for diet quality *, Viçosa, Brazil, 2009. Variable
BHEI-R Best diet quality
BHEI-R Intermediate diet quality n (%)
OR (IC 95%)
n (%)
OR (IC 95%)
Gender Male
92 (44,5)
1,0
81 (39,1)
1,0
Female
115 (55,6)
1,6 (1,1-2,4)
126 (60,9)
2,0 (1,4-3,0)
Age range (years) 60-74
147 (71,0)
1,0
147 (71,0)
1,0
≥ 75
60 (29,0)
0,8 (0,6-1,3)
60 (29,0)
38 (18,4)
1,0
63 (30,4)
1,0
135 (65,2)
1,2 (0,7-2,0)
125 (60,4)
2,0 (1,1-3,6)
34 (16,4)
1,6 (0,8-3,2)
19 (9,2)
4,9 (2,4-9,8)
0,8 (0,6-1,3)
Education Never studied Early grades of elementary school Latest grades of elementary or higher Individual income (quartiles) Q4 (≥ R$ 1476.25)
51 (24,8)
1,0
61 (30,2)
1,0
Q3 (R$ 522.50 – 1476.24)
54 (26,2)
0,9 (0,5-1,7)
50 (24,8)
0,6 (0,3-1,2)
Q2 (R$ 465.00 – 522.49)
78 (37,9)
1,1 (0,5-2,2)
65 (32,2)
0,9 (0,5-1,8)
Q1 (R$ 0 – 464.99)
23 (11.2) 1,3 (0,7-2,6) 26 (12,9) 1,3 (0,7-2,6) IC95%: 95% confidence interval; OR: odds ratio; PC: waist circumference. BHEI-R: Brazilian Healthy Eating Index–Revised. *For this analysis, the reference category was the lowest for the tertile with BHEI-R of “Poor diet quality”.
Table 4 - Univariate analysis of the association between the variables health, use of health services, lifestyle, waist circumference and diet quality*, Viçosa, Brazil, 2009. Variable
BHEI-R Intermediate diet quality n (%) OR (IC 95%)
BHEI-R Best diet quality n (%) OR (IC 95%)
Functional capability Adequate
174 (84,1)
1,0
169 (82,4)
1,0
Inadequate
33 (16,0)
1,1 (0,6-1,9)
36 (17,6)
1,2 (0,7-2,1)
Difficulty to eat No
189 (91,3)
1,0
191 (92,3)
1,0
18 (8,7)
0,9 (0,5-1,8)
16 (7,7)
0,8 (0,4-1,7)
Difficulty to prepare food No
126 (86,3)
1,0
112 (81,8)
1,0
Yes
20 (13,7)
0,8 (0,4-1,6)
25 (18,2)
1,1 (0,6-2,2)
Medical appointments (past year) None
Yes
17 (8,2)
1,0
5 (2,4)
1,0
1 to 5
148 (71,5)
1,4 (0,7-2,7)
156 (75,7)
4,9 (1,8-13,4)
≥6
42 (20,3)
1,5 (0,7-3,2)
45 (21,8)
5,4 (1,9-15,7)
Number of disorders ≤4
138 (66,7)
1.0
104 (50,2)
1,0
≥5
69 (33,4)
1,1 (0,8-1,7)
103 (49,8)
2,2 (1,5-3,4)
History of diabetes mellitus No
169 (81,6)
1,0
131 (63,3)
1,0
Yes
38 (18,4)
1,7 (1,0-3,0)
76 (36,7)
4,4 (2,6-7,3)
History of dyslipidemia No
96 (46,4)
1,0
62 (29,9)
1,0
Yes
110 (53,4)
1,3 (0,9-1,9)
145 (70,0)
2,6 (1,8-3,9)
History of high blood pressure No
54 (26,1)
1,0
39 (18,8)
1,0
Yes
153 (73,9)
1,0 (0,6-1,5)
168 (81,2)
1,5 (0,9-2,4)
Polypharmacy No
140 (67,6)
1,0
104 (50,2)
1,0
Yes
67 (32,4)
1,3 (0,9-2,1)
103 (49,8)
2,8 (1,8-4,2)
Alcohol consumption Never drank
73 (35,3)
1,0
69 (33,5)
1,0
No but has drunk before
65 (31,4)
0,8 (0,5-1,4)
70 (34,0)
1,0 (0,6-1,6)
Yes
69 (33,4)
0,9 (0,6-1,5)
67 (32,5)
0,9 (0,6-1,5)
Smoking Never smoked
115 (55,8)
1,0
122 (59,2)
1,0
No but has smoked before
73 (35,4)
1,0 (0,6-1,5)
62 (30,1)
0,8 (0,5-1,2)
Yes
18 (8,7)
0,6 (0,3-1,2)
22 (10,7)
0,7 (0,4-1,3)
Reduction of food intake No
184 (88,9)
1,0
182 (87,9)
1,0
Yes
23 (11,1)
0,7 (0,4-1,3)
25 (12,1)
0,8 (0,5-1,4)
PC (tertiles)
Lower than the first tertile
54 (26,7)
1,0
65 (32,2)
1,0
Between the first and second tertiles
80 (40,8)
2,4 (1,5-3,9)
65 (33,2)
1,6 (1,0-2,7)
Higher our equal to second tertile
66 (33,3) 1,6 (0,9-2,5) 67 (33,8) 1,3 (0,8-2,1) IC95%: 95% confidence interval; OR: odds ratio; PC: waist circumference. BHEI-R: Brazilian Healthy Eating Index–Revised. *For this analysis, the reference category was the lowest for the tertile with BHEI-R of “Poor diet quality”.
Table 5 - Results of the multivariate analysis of the association between sociodemographic conditions, health status, use of health services, lifestyle, waist circumference and diet quality *, Viçosa, Brazil, 2009. BHEI-R Intermediate diet quality OR (IC 95%)
BHEI-R Best diet quality OR (IC 95%)
Gender Male
1,0
1,0
Female
1,8 (1,1-2,9)
1,7 (1,0-2,9)
Variable
Education Never studied
1,0
1,0
Early grades of elementary school
1,4 (0,8-2,4)
3,1 (1,5-6,1)
Latest grades of elementary or higher
2,0 (0,9-4,1)
10,5 (4,5-24,5)
Medical appointments (past year) None
1,0
1,0
1 to 5
1,2 (0,6-2,6)
≥6
1,2 (0,5-2,9)
3,7 (1,0-13,1) 3,2 (0,8-12,5)
1,0
1,0
1,4 (0,8-2,6)
4,1 (2,3-7,5)
1,0
1,0
1,1 (0,7-1,7)
1,6 (1,0-2,6)
1,0
1,0
0,8 (0,5-1,3)
1,1 (0,7-2,0)
1,0
1,0
1,1 (0,7-1,8)
1,8 (1,1-3,0)
History of diabetes mellitus No Yes History of dyslipidemia No Yes History of high blood pressure No Yes Polypharmacy No Yes Alcohol consumption Never drank
1,0
1,0
No but has drunk before
1,3 (0,8-2,0)
0,8 (0,5-1,4)
Yes
0,9 (0,4-1,9)
1,4 (0,7-2,9)
1,0
1,0
2,5 (1,5-4,2)
1,6 (0,9-2,7)
PC (tertiles) Lower than the first tertile Between the first and second tertiles Higher our equal to second tertile
1,5 (0,9-2,4) 0,9 (0,5-1,5) IC95%: 95% confidence interval; OR: odds ratio; PC: waist circumference. BHEI-R: Brazilian Healthy Eating Index–Revised. *For this analysis, the reference category was the lowest for the tertile with BHEI-R of “Poor diet quality”.